Assessing the spatiotemporal variability of SMAP soil moisture accuracy in a deciduous forest region

Authors: ABDELKADER, M. - Stevens Institute Of Technology; TEMIMI, M. - Stevens Institute Of Technology; COLLIANDER, A. - Jet Propulsion Laboratory; Cosh, Michael; KELLY, V. - Cary Institute Of Ecosystem Studies; LAKHANKAR, T. - Collaborator; FARES, A. - Prairie View A & M University

Submitted to: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/5/2022
Publication Date: 7/11/2022
Citation: Abdelkader, M., Temimi, M., Colliander, A., Cosh, M.H., Kelly, V., Lakhankar, T., Fares, A. 2022. Assessing the spatiotemporal variability of SMAP soil moisture accuracy in a deciduous forest region. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 14(14). https://doi.org/10.3390/rs14143329.
DOI: https://doi.org/10.3390/rs14143329

Interpretive Summary: Soil moisture monitoring from remote sensing for the eastern portion of the United States is complicated by the presence of large heterogeneous forest stands. Most satellites which monitor soil moisture are microwave based and the signal is confounded by high biomass vegetation. A revision to the retrieval algorithm for forested and forest/agriculture mixed landscapes is necessary if these products are going to be valid for large portions of the U.S. A study was conducted on a soil moisture network deployed in Dutchess County around the Cary Institute of Ecosystem Studies in Millbrook, New York. This network was supporting the Soil Moisture Active Passive (SMAP) Validation Experiment 2019-2022. It was determined that for the high biomass periods during the summer, the agreement between the network and the SMAP product was the worst, while during winter months, the agreement was closer to target mission metrics. In the future, further refinement of the algorithm with better biomass information will hopefully improve the performance.

Technical Abstract: The goal of this study is to assess the temporal variability of the performance of SMAP soil moisture retrievals throughout the seasons as surface conditions change. In situ soil moisture observations from a network deployed in Millbrook, New York, between 2019 and 2021 are used. The network comprises 25 stations distributed across a 33-km SMAP pixel with a predominant forest land cover. Four upscaling methods are used in this study: arithmetic average, Voronoi diagram, topographic wetness index, and land cover weighted average. The used in situ soil moisture observations were collected between 6:00 and 7:00 am, local time and the arithmetic mean soil moisture was calculated. The SMAP soil moisture was evaluated against the upscaled in situ measurements using the Percent Bias (PB) Root-Mean-Squared Difference (RMSD), the Mean Difference (MD), and the unbiased Root-Mean-Squared Difference (ubRMSD). The consistency of the temporal variability of SMAP soil moisture data resulting from the four upscaling methods was analyzed. The results revealed that SMAP retrievals are systematically higher than in situ observations during the different seasons. Results indicated that the highest performance of SMAP soil moisture retrievals was during the month of November with the ubRMSD ranging between 0.016 and 0.031 m3 m-3 for the four upscaling methods which can be attributed to lower vegetation density, due to seasonal transition. The agreement with in situ observations degrades during January-March with ubRMSD values above 0.04 m3 m-3, reaching 0.06 m3 m-3 in March which can be attributed to the non-reliability of in situ observations due to freeze\thaw transition and the challenging determination of the soil effective temperature. The ubRMSD was also higher than 0.04 m3 m-3 in the months of April-June which can be explained by the introduced vegetation effect during the growth season. These findings were consistent among all the upscaling methods. The average ubRMSD over the study period was 0.065 m3 m-3, which fell short of meeting the mission performance target. This study corroborates the need to enhance SMAP retrieval over forest sites while taking into account the temporal variability of soil moisture.